Scholarlink

<p>The objectives of this paper are (1) revealing spatial patterns of residential PV adoption in Florida; (2) identify significant influencing factors and capture their global effects; (3) map spatially varying magnitude and direction of factors; and (4) predict probability of PV adoption. Cartographic visualization and exploratory analysis of data PV adoption in Florida from 2007 to 2010 after a pilot incentives program found that 54% of residential PV systems were in East Central, Southwest and Tampa Bay. The Log-Gaussian Cox (LGCP) point process model found that s<i>ignificant positive factors</i> affecting PV adoption are house age/density/value, education level, income, and retirement, and <i>significant negative factors</i> are health and mortgage. The geographically weighted regression (GWR) model found that i<i>ncome</i> effect on PV adoption is positive almost all over the state with strongest effect in East Central. <i>Race</i> is negative all over the state with strongest effect in North Central, Northeast, and Southwest. <i>Density</i> is negative all over the state with strongest effect in Apachee. All other variables show effects of varying magnitude and direction across the state. The deep neural network (DNN) model found that areas most probable to adopt PV are along the Atlantic coast, Gulf coast from Tampa Bay to Southwest, Norwest Florida coast, and East Central. The findings help government and PV system businesses target households and geographic regions of certain socioeconomic status for equitable incentive support and marketing. Our study contributes to literature by testing a suite of spatial analysis and modeling methods, including two unique models (LGCP and DNN) no prior studies have used for PV adoption modeling.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Spatial pattern analysis and modeling of residential adoption of photovoltaic system in Florida

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月19日

查看全文
  • Zhiyong Hu,
  • James Hu

摘要

The objectives of this paper are (1) revealing spatial patterns of residential PV adoption in Florida; (2) identify significant influencing factors and capture their global effects; (3) map spatially varying magnitude and direction of factors; and (4) predict probability of PV adoption. Cartographic visualization and exploratory analysis of data PV adoption in Florida from 2007 to 2010 after a pilot incentives program found that 54% of residential PV systems were in East Central, Southwest and Tampa Bay. The Log-Gaussian Cox (LGCP) point process model found that significant positive factors affecting PV adoption are house age/density/value, education level, income, and retirement, and significant negative factors are health and mortgage. The geographically weighted regression (GWR) model found that income effect on PV adoption is positive almost all over the state with strongest effect in East Central. Race is negative all over the state with strongest effect in North Central, Northeast, and Southwest. Density is negative all over the state with strongest effect in Apachee. All other variables show effects of varying magnitude and direction across the state. The deep neural network (DNN) model found that areas most probable to adopt PV are along the Atlantic coast, Gulf coast from Tampa Bay to Southwest, Norwest Florida coast, and East Central. The findings help government and PV system businesses target households and geographic regions of certain socioeconomic status for equitable incentive support and marketing. Our study contributes to literature by testing a suite of spatial analysis and modeling methods, including two unique models (LGCP and DNN) no prior studies have used for PV adoption modeling.